Skip to contents

Spatialized trees and forest stands metrics with BIOMASS

Arthur Bailly

2024-12-02

Source: vignettes/Vignette_spatialized_trees_and_forest_stands_metrics.Rmd
Vignette_spatialized_trees_and_forest_stands_metrics.Rmd

/!\ VIGNETTE IN PROGRESS /!\

Overview

BIOMASS enables users to manage their plots by :

  • calculating the projected/geographic coordinates of the plot’s corners and trees from the relative coordinates (or local coordinates, ie, those of the field)

  • visualizing the plots

  • validating plot’s corner and tree coordinates with LiDAR data

  • dividing plots into subplots

  • summarising the desired information at subplot level

Required data

Two data frames are required for the rest :

  • a data frame of corner coordinates, containing at least :
    • the name of the plots if there are several plots
    • the coordinates of the plot’s corners in the geographic or projected coordinate system (the GPS coordinates or another projected coordinates)
    • the coordinates of the plot’s corners in the relative coordinate system (the local or field coordinates)

In this vignette, we will use the Nouragues dataset as an exemple :

corner_coord <- read.csv(system.file("external", "Coord.csv",package = "BIOMASS", mustWork = TRUE))

kable(head(corner_coord), digits = 5, row.names = FALSE, caption = "Head of the table corner_coord")
Head of the table corner_coord
Plot Corners Lat Long xRel yRel xProj yProj
NB1 NB1_SE 4.06692 52.68883 100 0 687482.0 449720.8
NB1 NB1_SE 4.06694 52.68883 100 0 687480.9 449722.6
NB1 NB1_SE 4.06694 52.68884 100 0 687482.8 449722.2
NB1 NB1_SE 4.06692 52.68882 100 0 687480.7 449720.7
NB1 NB1_SE 4.06695 52.68883 100 0 687481.5 449723.5
NB1 NB1_SE 4.06695 52.68884 100 0 687483.1 449723.5

Note that both Lat/Long and xProj/yProj coordinates are included in this dataset but only one of these coordinate systems is needed.

  • a data frame of trees coordinates, containing at least :
    • the name of the plots if there are several plots
    • the tree coordinates in the relative coordinate system (the local/field one)
    • the desired information about trees, such as diameter, wood density, height, etc. (see BIOMASS vignette)
trees <- read.csv(system.file("external", "NouraguesPlot.csv",package = "BIOMASS", mustWork = TRUE))

kable(head(trees), digits = 3, row.names = FALSE, caption = "Head of the table trees")
Head of the table trees
plot xRel yRel D WD H
NB1 1.30 4.7 11.459 0.643 12
NB1 2.65 4.3 11.618 0.580 16
NB1 4.20 6.9 83.875 0.591 40
NB1 5.90 4.7 14.961 0.568 18
NB1 6.40 4.1 36.765 0.530 27
NB1 13.50 2.3 13.528 0.409 20

Checking plot’s coordinates

The user is faced with two situations :

  • The GPS coordinates of the plot corners are considered very accurate or enough measurements have been made to be confident in the accuracy of their mean. In this case, the shape of the plot measured on the field will follow the GPS coordinates of the plot corners when projected into the projected/geographic coordinate system. See 3.1.1

  • Too few measurements of the GPS coordinates of plot corners have been taken and/or are not reliable. In this case, the shape of the plot measured on the field is considered to be accurate and the GPS corner coordinates will be recalculated to fit the shape and dimensions of the plot (the projected coordinates fit the relative coordinates). See 3.1.2

In both case, the use of the check_plot_coord() function is recommended as a first step.

Checking the corners of the plot

The check_plot_coord() function handles both situations using the trust_GPS_corners argument (= TRUE or FALSE).

You can give either GPS coordinates (with the longlat argument) or another projected coordinates (with the proj_coord argument) for the corner coordinates.

If we rely on the GPS coordinates of the corners

If enough coordinates have been recorded for each corner (for more information, see the CEOS good practices protocol, section A.1.3.1 ), you will need to provide the corner names via the corner_ID argument. In this case, each coordinates will be averaging by corner, resulting in 4 reference coordinates. The function can also detect and remove GPS outliers using the rm_outliers and max_dist arguments.

If only 4 GPS measurements have been taken with a high degree of accuracy (by a geometer, for example), or if you already have averaged your measurements by yourself, you can supply these 4 GPS coordinates to the function.

check_plot_trust_GPS <- check_plot_coord(
  longlat = corner_coord[c("Long", "Lat")],
  # Or, if exists : proj_coord = corner_coord[c("xProj", "yProj")], 
  rel_coord = corner_coord[c("xRel", "yRel")],
  trust_GPS_corners = T, corner_ID = corner_coord$Corners,
  draw_plot = TRUE,
  max_dist = 10, rm_outliers = TRUE )
#> Loading required namespace: proj4

The two blue arrows represent the relative coordinate system when projected into the projected coordinate system.

If we rely on the shape of the plot measured on the field

Let’s degrade the data to simulate the fact that we only have 8 GPS coordinates that we don’t trust.

degraded_corner_coord <- corner_coord[c(1:2,11:12,21:22,31:32),]
  
check_plot_trust_field <- check_plot_coord(
  longlat = degraded_corner_coord[, c("Long", "Lat")],
  # OR proj_coord = corner_coord[c("xProj", "yProj")], 
  rel_coord = degraded_corner_coord[, c("xRel", "yRel")],
  trust_GPS_corners = F,
  draw_plot = TRUE )

We can see that the corners of the plot do not correspond to the GPS measurements. In fact, they correspond to the best compromise between the shape and dimensions of the plot and the GPS measurements.

Retrieving tree projected coordinates

Tree coordinates are almost always measured in the relative (local/field) coordinate system. To retrieve them in the projected system, you can supply their relative coordinates using the tree_df and tree_coords arguments.

check_plot_trust_GPS <- check_plot_coord(
  longlat = corner_coord[c("Long", "Lat")],
  # OR proj_coord = corner_coord[c("xProj", "yProj")], 
  rel_coord = corner_coord[c("xRel", "yRel")],
  trust_GPS_corners = T, corner_ID = corner_coord$Corners,
  draw_plot = TRUE,
  max_dist = 10, rm_outliers = TRUE,
  tree_df = trees, tree_coords = c("xRel","yRel") )


check_plot_trust_field <- check_plot_coord(
  longlat = degraded_corner_coord[, c("Long", "Lat")],
  # OR proj_coord = corner_coord[c("xProj", "yProj")], 
  rel_coord = degraded_corner_coord[, c("xRel", "yRel")],
  trust_GPS_corners = F,
  draw_plot = TRUE,
  tree_df = trees, tree_coords = c("xRel","yRel") )

Note the difference in corner and tree positions between the two situations.

The tree coordinates can be obtained via the $tree_proj_coord output.

trees[c("x_proj","y_proj")] <- check_plot_trust_GPS$tree_proj_coord

You can also grep and modify the plot via the $plot_design output which is a ggplot object. For example, to change the plot title :

plot_to_change <- check_plot_trust_GPS$plot_design
plot_to_change <- plot_to_change + ggtitle("A custom title")
plot_to_change

If you want to retrieve the GPS coordinates of the trees in a longitude/latitude format, use this code:

tree_GPS_coord <- as.data.frame( proj4::project(check_plot_trust_GPS$tree_proj_coord, proj = check_plot_trust_GPS$UTM_code, inverse = TRUE) )

Integrating LiDAR data

WORK IN PROGRESS

Dividing plots

Dividing plots into several rectangular sub-plots is performed using the divide_plot() function. This function takes the relative coordinates of the 4 corners of the plot to divide it into a grid (be aware that the plot must be rectangular in the relative coordinates system, ie, have 4 right angles) :

subplots <- divide_plot(
  rel_coord = check_plot_trust_GPS$corner_coord[c("x_rel","y_rel")],
  proj_coord = check_plot_trust_GPS$corner_coord[c("x_proj","y_proj")],
  grid_size = 25 # or c(25,25)
  )

kable(head(subplots, 10), digits = 1, row.names = FALSE, caption = "Head of the divide_plot() returns")
Head of the divide_plot() returns
subplot_id x_rel y_rel x_proj y_proj
subplot_0_0 0 0 687372.8 449717.9
subplot_0_0 25 0 687400.1 449719.0
subplot_0_0 25 25 687401.3 449744.7
subplot_0_0 0 25 687374.9 449743.8
subplot_1_0 25 0 687400.1 449719.0
subplot_1_0 50 0 687427.4 449720.0
subplot_1_0 50 25 687427.7 449745.6
subplot_1_0 25 25 687401.3 449744.7
subplot_2_0 50 0 687427.4 449720.0
subplot_2_0 75 0 687454.6 449721.0

If you don’t have any projected coordinates, just let proj_coord = NULL.

For the purposes of summarising and representing subplots (coming in the next section), the function returns the coordinates of subplot corners but can also assign to each tree its subplot with the tree_df and tree_coords arguments :

trees$AGB <- computeAGB(trees$D, trees$WD, H = trees$H)

subplots <- divide_plot(
  rel_coord = check_plot_trust_GPS$corner_coord[c("x_rel","y_rel")],
  proj_coord = check_plot_trust_GPS$corner_coord[c("x_proj","y_proj")],
  grid_size = 25, # or c(50,50)
  tree_df = trees, tree_coords = c("xRel","yRel")
  )

The function now returns a list containing :

  • sub_corner_coord : coordinates of subplot corners as previously

  • tree_df : the tree data-frame with the subplot_id added in last column

kable(head(subplots$tree_df), digits = 1, row.names = FALSE, caption = "Head of the divide_plot()$tree_df returns")
Head of the divide_plot()$tree_df returns
plot xRel yRel D WD H x_proj y_proj AGB subplot_id
NB1 1.3 4.7 11.5 0.6 12 687374.6 449722.8 0.1 subplot_0_0
NB1 2.6 4.3 11.6 0.6 16 687376.0 449722.5 0.1 subplot_0_0
NB1 4.2 6.9 83.9 0.6 40 687377.9 449725.2 8.4 subplot_0_0
NB1 5.9 4.7 15.0 0.6 18 687379.6 449723.0 0.1 subplot_0_0
NB1 6.4 4.1 36.8 0.5 27 687380.1 449722.4 1.0 subplot_0_0
NB1 13.5 2.3 13.5 0.4 20 687387.7 449720.9 0.1 subplot_0_0

The function can handle as many plots as you wish, using the corner_plot_ID and tree_plot_ID arguments :

multiple_plots <- rbind(check_plot_trust_GPS$corner_coord[c("x_rel","y_rel","x_proj","y_proj")], check_plot_trust_GPS$corner_coord[c("x_rel","y_rel","x_proj","y_proj")])
multiple_plots$plot_ID = rep(c("NB1","NB2"),e=4)
multiple_trees <- rbind(trees,trees)
multiple_trees$plot = rep(c("NB1","NB2"),e=nrow(trees))

multiple_subplots <- divide_plot(
  rel_coord = multiple_plots[c("x_rel","y_rel")],
  proj_coord = multiple_plots[c("x_proj","y_proj")],
  grid_size = 50,
  corner_plot_ID = multiple_plots$plot_ID,
  tree_df = multiple_trees,
  tree_coords = c("xRel","yRel"),
  tree_plot_ID = multiple_trees$plot
)

Summarising tree metrics at subplot level

Once you’ve applied the divide_plot() function with a tree_df argument, you can summarise any tree metric present in divide_plot()$tree_df at the subplot level with the subplot_summary() function.

# AGB metric was added before applying the divide_plot() function, but it could have been added afterwards with : 
# subplots$tree_df$AGB <- computeAGB(trees$D, trees$WD, H = trees$H)

subplot_metric <- subplot_summary(subplots = subplots,
                                  value = "AGB") # AGB is the tree metric

By default, the function sums the metric per subplot but you can request any valid function with the fun argument :

subplot_metric <- subplot_summary(subplots = subplots,
                                  value = "AGB",
                                  fun = quantile, probs = 0.9)

The output of the function is a list containing :

  • $tree_summary : a summary of the metric per subplot

  • $polygon : an sf object containing a simple feature collection of the subplot’s polygon

  • $plot_design : (if draw_plot=T) a ggplot object that can easily be modified

Customizing the ggplot

Here are some examples to custom the ggplot of the subplot_summary() function :

subplot_metric <- subplot_summary(subplots = subplots,
                                  value = "AGB") 

custom_plot <- subplot_metric$plot_design
# Change the title and legend :
custom_plot + 
  labs(title = "Nourragues plot" , fill="Sum of AGB per hectare")

# Display trees : 
custom_plot + 
  geom_point(data=trees, mapping = aes(x = x_proj, y = y_proj), shape=3) 

# Display trees with diameter as size point : 
custom_plot + 
  geom_point(data=trees, mapping = aes(x = x_proj, y = y_proj, size = D), shape=1) +
  labs(size = "Tree diameter (m)", fill = "Sum of AGB per hectare") 

# Display trees with diameter as size point and height as color point (and a smaller legend on the right) : 
custom_plot + 
  geom_point(data=trees, mapping = aes(x = x_proj, y = y_proj, size = D, col = H), shape=1) +
  labs(size = "Tree diameter (m)", col = "Tree size (m)", fill = "Sum of AGB per hectare") +
  scale_color_gradientn(colours = c("white","black")) +
  theme(legend.position = "right", legend.key.size = unit(0.5, 'cm'))

Exporting as a shapefile 

subplot_polygons <- subplot_metric$polygon 
# sf::st_crs(subplot_polygons)

Dealing with multiple plots 

multiple_subplot_metric <- subplot_summary(
  subplots = multiple_subplots,
  value = "AGB")